crypto_trader/docs/user_manual/faq.md

# Frequently Asked Questions

Common questions about Crypto Trader.

## General Questions

### What is Crypto Trader?

Crypto Trader is a comprehensive web-based cryptocurrency trading platform for trading, backtesting, and portfolio management. It features a modern React frontend with a FastAPI backend, supporting multiple exchanges, real-time trading, paper trading, and advanced analytics.

### Is Crypto Trader free?

[Answer depends on your licensing - update as needed]

### What operating systems are supported?

The web interface works on any operating system with a modern browser:
- **Backend**: Linux, macOS, Windows (Python 3.11+)
- **Frontend**: Any OS with Chrome, Firefox, Safari, or Edge
- **Recommended**: Linux (Bluefin Linux) or macOS for development

### What Python version is required?

Python 3.11 or higher is required.

## Installation Questions

### How do I install Crypto Trader?

See the [Getting Started](getting_started.md) guide for installation instructions.

### Can I run from source?

Yes, see the [Getting Started](getting_started.md) guide. You'll need to run both the backend (FastAPI) and frontend (React) servers.

### Do I need to install TA-Lib separately?

TA-Lib is bundled in the AppImage. For source installation, you may need to install the TA-Lib C library separately.

## Trading Questions

### Is paper trading safe?

Yes, paper trading uses virtual funds and doesn't risk real money. It's perfect for testing strategies.

### How do I switch from paper trading to live trading?

1. Configure exchange API keys with trading permissions
2. Disable paper trading mode in settings
3. Start with small position sizes
4. Monitor closely

### What exchanges are supported?

Currently Coinbase is supported. The framework supports adding additional exchanges.

### Can I trade futures?

Yes, futures and leverage trading are supported. See the trading guide for details.

## Strategy Questions

### How do I create a strategy?

See the [Strategy Guide](strategies.md) for detailed instructions.

### Can I use multiple strategies at once?

Yes, you can run multiple strategies simultaneously on different symbols or timeframes.

### How do I optimize strategy parameters?

Use the backtesting optimization features. See the [Backtesting Guide](backtesting.md#parameter-optimization).

### Can I create custom strategies?

Yes, see the [Developer Guide](../developer/creating_strategies.md) for instructions.

## Backtesting Questions

### How accurate is backtesting?

Backtesting includes realistic features like slippage and fees, but past performance doesn't guarantee future results.

### How much historical data do I need?

At least 6-12 months of data is recommended for reliable backtesting.

### Can I backtest multiple strategies?

Yes, you can backtest multiple strategies and compare results.

## Autopilot Questions

### What's the difference between Pattern-Based and ML-Based Autopilot?

**Pattern-Based Autopilot** uses technical chart pattern recognition combined with sentiment analysis. It's transparent, explainable, and works immediately without training data. **ML-Based Autopilot** uses machine learning to select the best strategy based on market conditions. It's adaptive and learns from historical performance but requires training data.

See the [Trading Guide](trading.md#autopilot-modes) for detailed comparison.

### Which autopilot mode should I use?

- **Use Pattern-Based** if you want transparency, understand technical analysis, prefer explainable decisions, or want immediate setup.
- **Use ML-Based** if you want adaptive decisions, have historical trading data, don't need to understand every decision, or want maximum optimization.

See the [decision guide](trading.md#choosing-the-right-autopilot-mode) for more details.

### Can I switch between modes?

Yes, you can switch between modes at any time. Simply stop the current autopilot, select the desired mode, and start it again. You cannot run both modes simultaneously for the same symbol.

### What is auto-execution and should I enable it?

Auto-execution automatically executes trades based on autopilot signals. It's available for both modes and can be enabled/disabled independently.

**Enable auto-execution if**:
- You've tested the autopilot in paper trading mode
- You trust the autopilot's signals
- You've set appropriate risk limits
- You can monitor trades regularly

**Don't enable auto-execution if**:
- You're new to the autopilot
- You haven't tested it thoroughly
- You want to review signals before executing
- You're trading with real money and want manual control

### How do I know which mode is running?

The Dashboard shows the active mode in the status chip (e.g., "AutoPilot Active (pattern)"). You can also check the Autopilot Configuration section to see the selected mode.

### Can I use both modes simultaneously?

No, you can only run one autopilot mode at a time per symbol. However, you can run different modes on different symbols.

### What happens if I switch modes while autopilot is running?

You must stop the current autopilot before switching modes. The system will prevent starting a new mode while another is running for the same symbol.

### Does Pattern-Based Autopilot require training data?

No, Pattern-Based Autopilot works immediately without any training data. It uses predefined pattern recognition rules and sentiment analysis.

### Does ML-Based Autopilot always need training data?

Yes, ML-Based Autopilot requires training data to build its model. However, the system can bootstrap training data from historical backtests if available. You can also manually trigger model training from the Dashboard.

### How accurate are autopilot signals?

Signal accuracy depends on market conditions and the selected mode:
- **Pattern-Based**: Accuracy depends on pattern clarity and sentiment alignment
- **ML-Based**: Accuracy depends on model training quality and market regime match

Past performance doesn't guarantee future results. Always use risk management and test thoroughly in paper trading mode.

## Data Questions

### Where is data stored?

Data is stored in `~/.local/share/crypto_trader/`:
- Database: `trading.db`
- Historical data: `historical/`
- Logs: `logs/`

### How much storage do I need?

Storage depends on:
- Historical data retention
- Number of symbols tracked
- Log retention settings

Typically 1-5GB is sufficient.

### Can I use PostgreSQL instead of SQLite?

Yes, PostgreSQL is supported for advanced users. See the [Configuration Guide](configuration.md#database-options).

## Security Questions

### Are my API keys secure?

Yes, API keys are encrypted before storage using industry-standard encryption.

### Can I use read-only API keys?

Yes, read-only keys are recommended for backtesting and data collection.

### Where are API keys stored?

API keys are stored encrypted in the SQLite database.

## Performance Questions

### Why is the application slow?

Possible causes:
- Large historical datasets
- Many active strategies
- Insufficient system resources
- Database optimization needed

See [Troubleshooting](troubleshooting.md#performance-issues) for solutions.

### How can I improve performance?

- Reduce data retention
- Limit active strategies
- Optimize database
- Increase system resources

## Technical Questions

### Can I contribute to the project?

Yes! See the [Developer Guide](../developer/contributing.md) for contribution guidelines.

### How do I report bugs?

Create an issue with:
- Error description
- Steps to reproduce
- System information
- Log files

### Is there an API?

Yes, see the [API Documentation](../api/index.html) for details.

## Still Have Questions?

- Check the [User Manual](README.md)
- Review [Troubleshooting](troubleshooting.md)
- Consult [API Documentation](../api/index.html)
- See [Developer Guide](../developer/README.md)